用难样本重采样缓解分类偏差，提升模型公平性。
Reducing Class Bias In Data-Balanced Datasets Through Hardness-Based Resampling
- 基于类别学习难度估计，动态调整数据重采样策略。
- 在CIFAR-10上降低32%召回差距，优于传统频率重采样。
- 适合关注模型公平性与数据质量优化的研究者。
类别偏差（即类别间性能差异）通常归因于数据不平衡，并通过频率重采样解决。然而我们发现，即使在完全平衡的数据集上，显著的偏差依然存在，证明仅靠类别频率无法解释模型性能不均。本文从类别学习难度视角出发，提出硬度感知重采样（HBR），利用难度估计指导数据选择。为更全面评估效果，引入补全球体指标的间隙与分散度测量。实验表明，HBR显著减少召回差距：在CIFAR-10上降低32%，在CIFAR-100上降低16%，优于标准频率重采样。进一步发现，使用先进扩散模型中最难样本进行选择，可比随机选取带来更好的公平性结果。研究证明，单纯数据平衡不足以缓解类别偏差，需转向硬度感知方法。
原文摘要 · Abstract (English)
Class-bias, that is class-wise performance disparities, is typically attributed to data imbalance and addressed through frequency-based resampling. However, we demonstrate that substantial bias persists even in perfectly balanced datasets, proving that class frequency alone cannot explain unequal model performance. We investigate these disparities through the lens of class-level learning difficulty and propose Hardness-Based Resampling (HBR), a strategy that leverages hardness estimates to guide data selection. To better capture these effects, we introduce an evaluation protocol that complements global metrics with gap- and dispersion-based measures. Our experiments show that HBR significantly reduces recall gaps, by up to 32% on CIFAR-10 and 16% on CIFAR-100, outperforming standard frequency-based resampling. We further show that we can improve fairness outcomes by selectively using the hardest samples from a state-of-the-art diffusion model, rather than randomly selecting them. These findings demonstrate that data balance alone is insufficient to mitigate class bias, necessitating a shift toward hardness-aware approaches.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。